So the purge valve should be connected to it's electrical connector, but the evap pipe off of it so it sucks in fresh air? Does it matter where I disconnect the evap, at the purge valve or at the canister?

Would would cause the high fuel vapor? Flooded charcoal canister or stuck closed vent valve?

I believe it did. Driving two of the exact same model and switching between them often I can't remember which I was in at the time. But if I remember correctly I had trouble gassing up with the pump stopping, then driving home and the issue showing up about 10 minutes after refueling.

 

When the purge valve opens in normal operation it sucks fresh air in through vent solenoid valve which is normally open and grabs some canister gas vapors on its way to the engine intake. I just want to confirm that taking the fuel tank and canister out of the circuit fixes your problem beyond just idle. So the normal fresh air intake but no gas vapor at all. If the pipe from the purge valve to the canister is disconnected at the canister then that should do it. The idea would be to then reattached the canister and check, then the gas tank with the original gas cap and check and then the vent solenoid to find out when the fuel trims take a large negative nose dive again. Then we address the problem. Hopefully the problem is not intermittent and it will allow us to isolate the cause. When you are back together with huge negative fuel trims you should also monitor fuel pressure to make sure that it does not go over spec and over pressurize the injectors. The regulator for instance is affected by changes in vacuum. The return line can also get blocked.

Yes flooded canister or a stuck vent valve and a replacement leak in the gas tank form the cap, filler neck, crack, hose, pressure sensor, etc. The normal process does not pull fumes from the tank as the only source of air once the purge valve places a vacuum on the system and only takes a small amount of vapor from the canister not the fumes from a flooded container.

The problem Starting after refueling does suggest a flooded canister and/or leaking gas cap and difficulty fueling suggests closed vent valve, blocked canister or lines.


George

 

I ran some tests.

I first did a 10 minute drive with the purge valve plugged in and evap line disconnected from it. I felt it was running great, very responsive to throttle. After the drive I parked and reconnected the evap line to the purge valve and so it was connected to the canister. The gas tank was still disconnected from the canister at this point. Throttle still felt great, very responsive and I could hold a pretty stable RPM. After another 10 minute drive like this I stopped and reconnected the gas tank to the canister. I immediately noticed a more sluggish throttle response just pulling out of the drive way. I only did an 8 minute drive for this test as I did not want to get too far out of the neighborhood with the longterms tanking. Note: The vent valve was disconnected from it's line and electrical connector for these tests.



I tried my best to overlay RPM with long term fuel trims. I've never been good at making data look pretty.

 

Your graphs are great, some of the best I have worked with here. Most people that come here are nowhere near as diligent as you are.

So the system needs to command a little more lean after the cannister hook up but the trims are still OK. I dont know if this indicates a flooded cannister or normal behavior. The cannister is not causing the worst case scenario on its own but may be contributing to another problem.

Connecting the tank without the evap solenoid obviously makes it worse but not as bad as before. Look at how the STFT jump straight to -18 until the LTFT catch up and start baking in the new condition.You dont have -30 trims at this point but -15 is still too high so the gas tank is involved in the problem. Was the gas cap attached or was the filler neck taped?

Lastly, lets see what happens when you hook the evap solenoid back up and plug it in so that everything is hooked back up, do you go back to -30 trims and an engine that barely runs? This is beginning to look like more than one problem. Hopefully we are getting close to doing something about this.

George

 

Thanks. I work in IT and really like finding the root of a problem instead of parts cannon. A lot cheaper too!

The gas cap was on. I wish I would have ran the last test with the tank connected a little longer. It really felt like long terms were going to go even more negative. I could definitely smell a difference in the exhaust. I'll hook everything back up like it's supposed to and go for a drive this evening and see what trims do.

 

After you hook everything back up and run it, then take the gas cap off and tape the filler neck again and run it one more time. That should be the end of the data gathering phase and we can start talking about what is probably going on here.


George

 

I had a thought.

When I ran the test with just the purge valve connected to a wide open canister, that essentially creates a vacuum leak, but the computer thinks it's sucking in fuel vapor, correct? Could this still be an issue somewhere else if the unmetered air it's sucking in leans out the extra fuel mixture that could be coming from somewhere else?

If it were the evap system, why does the computer not close the purge valve when it sees the engine getting too rich?

 

You go get that last set of graphs and I am going to grab one of my higher end scanners and see if I can command and read the purge valve status while watching fuel trims to get a deeper understanding of the purge valve cycles and conditions. Wont be until tomorrow though. The manual does not have an in depth description of the purge valve operation. That said, I thought that trims did not change much from purge valve open to the atmosphere commanded open and then closed? Unless of course its not really closing and we will talk about that after this last round of diagnostics.

George

 

I hooked the evap system back up with everything connected and plugged in and the gas cap on.




I then taped off the gas tank fill spout and went on another drive.

 

And how did it run during the drive portion in both tests?

George